Environmental News Network (ENN)  [Printer-friendly version]
November 20, 2006

RISK ASSESSMENT AT THE BIOMONITORING CROSSROADS

[Rachel's introduction: All the scientific advancements in risk
assessment of the past 20 years are becoming overshadowed by
biomonitoring, the capability to measure low levels of specific
toxicants (or their breakdown products) in humans, typically urine or
blood.]

By Dr. Gary Ginsberg

Nearly 25 years after the National Academies of Science codified the
practice of environmental risk assessment with their 1983 "Red
Book", this relatively young field now comes to a critical stage in
its maturation.

As the key scientific method that underlies toxics
regulations, the manner in which risk assessment is conducted is not
trivial. Since 1983 it has gone through many refinements, from
beginnings in crude, high dose animal tests, it has advanced to
increasingly more sophisticated procedures that tell us how toxicants
behave in animals and are likely to act in people. The newer models
forecast dose and toxicity in young children, pregnant women and those
whose metabolism may be deficient due to their unique genetic makeup.
But all these advances are becoming overshadowed by biomonitoring, the
capability to measure low levels of specific toxicants (or their
breakdown products) in humans, typically urine or blood.

Why is this such an important breakthrough and why is it putting risk
assessment at a precarious crossroads? First and foremost,
biomonitoring is teaching us that the old adage "What goes around,
comes around" is especially pertinent to toxic chemicals. It has
taught us this lesson many times as specialty chemicals that were
meant to stay in consumer products (e.g., flame retardants, Teflon
coating, plasticizers) end up inside us, likely from in utero onwards.

This greatly raises the stakes for risk assessment because we're not
just talking about possible exposures to theoretical receptors at some
point in the future, but exposures happening right now to real people
from whom we've just taken some blood. The October 2006 National
Geographic spread on biomonitoring highlights the finding of many
chemicals in one reporter's blood, with his levels of brominated flame
retardants much higher than typical. Certainly he and the rest of the
public want to know where all this exposure is coming from, what it
means to one's health, and why aren't there answers given that most
chemicals found have been used for decades. Everyone is taking a long
look at risk assessment for answers.

As pointed out in a new National Academy report on biomonitoring (NAS,
2006), there is good news -- the techniques exist. But the bad news is
that we are just starting to use them effectively, and even in those
cases, there is a question of whether we will believe the results.

Ironically, one of the main techniques for anwering the questions
raised by biomonitoring is... biomonitoring itself. Why use
biomonitoring just to search for chemicals? Why not also use it to
tell us whether people with lots of a chemical have more health
problems than those who don't. This is the arena of epidemiology, a
field that has many obstacles to overcome because people are much more
difficult to study than lab rats.

One of its main difficulties is determining the level of exposure in
study populations. Good biomonitoring data remove that mystery and
puts epidemiology squarely on the path of dose-response and risk
assessment. With more and more biomarkers available, epidemiologists
can test all sorts of hypotheses, and are now finding plausible links
between low dose chemical exposure and toxic effects in people. Risk
assessors are typically toxicologists, trained in the highly
artificial but well-controlled world of animal testing. Are risk
assessors ready to embrace the new wave of epidemiology that is
showing us that the controlled lab study may not tell us everything we
need to know about risk?

Hopefully the answer is yes. It was yes for lead and mercury, cases in
which biomonitoring in combination with decades of human testing
showed clear risks at low dose, driving lead from paint and gasoline
and causing us to warn pregnant women to cut down on fish consumption.

The new wave that's emerging is showing us that low levels of life's
common ingredients, from phthalates present in cosmetics to a newly
discovered dietary contaminant that's also in rockets and bombs
(perchlorate), have the potential to impair fetal development (Swan,
et al., 2005; Blount, et al., 2006). These findings are at low dose
levels, commonly experienced by average Americans.

The greater sensitivity of human testing may be related to the fact
that if we test enough people with biomonitoring-based epidemiology,
we will find a sensitive subgroup if it exists. Animals used in
toxicology testing are highly inbred so they are in another universe
when it comes to variability. CDC's new findings with perchlorate are
instructive: men and most women were insensitive to this chemical.
However, the subgroup of women with compromised iodine intake were
affected. These types of findings throw risk assessment into the
crossroads of biomonitoring -- does it embrace these new findings and
accept the caveats and uncertainties involved in human population
testing, or does it stick to the better controlled laboratory studies,
which as we are now finding out may be less sensitive than we had
originally believed.

The pendulum in recent years has been swinging towards epidemiology
for dose response information, limiting animal testing to demonstrate
plausibility and mechanism. The cases of phthalates and perchlorate
will show whether regulators are ready to fully accept the
biomonitoring-based risk assessment paradigm, testing their resolve
because it may mean more stringent regulation of cosmetics
(phthalates, as already being done in Europe) and the diet and
drinking water (perchlorate). The challenge for risk assessors will be
to use biomonitoring data responsibly to protect public health and not
let this new crossroads become bogged down by inaction and gridlock.

References

Blount, B. et al. (2006) Urinary Perchlorate and Thyroid Hormone
Levels in Adolescent and Adult Men and Women Living in the United
States Environ Health Perspect: doi:10.1289/ehp.9466. [Online 5
October 2006]

National Academy Science (2006) Human Biomonitoring for Environmental
Chemicals. National Academy Press.

Swan, S. et al. (2005) Decrease in anogenital distance among male
infants with prenatal phthalate exposure. Environ Health Perspect.
113:1056-61.

Dr. Gary Ginsberg is co-author of What's Toxic, What's Not (Berkley
Books, pub date 12/05/06) and is a toxicologist at the Connecticut
Department of Public Health. He is on the faculty of Yale University
and the University of Connecticut School of Medicine. He served on the
NAS Panel on Biomonitoring and currently serves on several other NAS
and EPA panels. For more information go to www.whatstoxic.com.

Copyright 2006 The Environmental News Network, Inc.